def dereplicate(fasta):
'''
Process the input fasta entries.
Populate a dictionary
{seq => {sample => counts}}. Use the special sample key None to
indicate the sum of all counts for that sequence.
'''
counts = {}
provenances = {}
for record in SeqIO.parse(fasta, 'fasta'):
seq = str(record.seq)
sample = util.strip_fasta_label(record.id)
if seq not in counts:
counts[seq] = 1
else:
counts[seq] += 1
if seq not in provenances:
provenances[seq] = {sample: 1}
elif sample not in provenances[seq]:
provenances[seq][sample] = 1
else:
provenances[seq][sample] += 1
return counts, provenances
def process_line(line):
fields = line.rstrip().split('\t')
query = util.strip_fasta_label(fields[0])
target = fields[1]
if target != no_hit:
write_out(query, target)
elif target == no_hit and save_no_hit:
write_out(query, 'no_hit')
def process_line(line):
fields = line.rstrip().split('\t')
seq = util.strip_fasta_label(fields[0])
hit_type = fields[1]
if hit_type == 'otu' and fields[2] == '*':
otu = fields[4]
elif hit_type == 'otu' and fields[2] != '*':
otu = fields[5]
elif hit_type == 'match':
otu = fields[4]
if hit_type != 'chimera':
output.write("{}: {}\n".format(seq, otu))
def b6_to_dict(b6, no_hit="*", save_no_hit=True):
'''from a blast6 mapping file, make a dict {query => target}'''
membership = {}
with open(b6) as f:
for line in f:
fields = line.split()
query = util.strip_fasta_label(fields[0])
target = fields[1]
if target != no_hit or save_no_hit:
membership[query] = target
return membership
def up_to_dict(up, keep_chimera=False):
'''from a uparse mapping file, make a dict {seq => otu}'''
membership = {}
with open(up) as f:
for line in f:
fields = line.split()
seq = util.strip_fasta_label(fields[0])
hit_type = fields[1]
otu = fields[4]
if hit_type == 'chimera':
if keep_chimera:
otu = 'chimera'
else:
# ignore this entry
continue
membership[seq] = otu
return membership
def b6_line_to_query_hit(line):
query, hit = line.rstrip().split("\t")[0:2]
query = util.strip_fasta_label(query)
return query, hit
